How to integrate siren activation function to cnn - machine-learning

Question

I was trying to implement an autoencoder with periodic activation function - siren Link. But I have found examples using only linear layer. In the paper they mentioned a custom Hypernetwork which uses CNN but i didn't understood how to use for my autoencoder.
Is it possible to implement siren for CNN
how to implement for my autoencoder code mentioned in the below. simply adding sin() as activation function doesn't work because the author mentioned a custom initialization and some other tricks to make the periodic activation work.
I am attaching the paper here - Link
class Autoencoder(nn.Module):
def __init__(self):
super(Autoencoder, self).__init__()
# encoder layers
self.enc1 = nn.Conv2d(1, 64, kernel_size=3, padding=1)
self.enc2 = nn.Conv2d(64, 32, kernel_size=3, padding=1)
self.enc3 = nn.Conv2d(32, 16, kernel_size=3, padding=1)
self.enc4 = nn.Conv2d(16, 8, kernel_size=3, padding=1)
self.pool = nn.MaxPool2d(2, 2)
# decoder layers
self.dec1 = nn.ConvTranspose2d(8, 8, kernel_size=3, stride=2)
self.dec2 = nn.ConvTranspose2d(8, 16, kernel_size=3, stride=2)
self.dec3 = nn.ConvTranspose2d(16, 32, kernel_size=2, stride=2)
self.dec4 = nn.ConvTranspose2d(32, 64, kernel_size=2, stride=2)
self.out = nn.Conv2d(64, 1, kernel_size=3, padding=1)
def forward(self, x):
# encode
x = F.relu(self.enc1(x))
x = self.pool(x)
x = F.relu(self.enc2(x))
x = self.pool(x)
x = F.relu(self.enc3(x))
x = self.pool(x)
x = F.relu(self.enc4(x))
x = self.pool(x) # the latent space representation
# decode
x = F.relu(self.dec1(x))
x = F.relu(self.dec2(x))
x = F.relu(self.dec3(x))
x = F.relu(self.dec4(x))
x = F.sigmoid(self.out(x))
return x
net = Autoencoder()
print(net)